rs2476601

Data file: ic_t1d_onengut_allresults.csv.

1. Build 37 method

  • Get corresponding region from immunochip-regions.txt –> chr1:113863087-114527968.
  • Get SNPs in this region from ic_t1d_onengut_allresults.csv file (694 SNPs).
  • Use aligned9.build37.RData to get LD matrix for these SNPs (drop 55 SNPs here which do not appear in this file –> 639 SNPs). Sigma <- ld(Xsub, Xsub, stat="R")
  • Proceed as normal.
## $Credset
##     name.rs        pp
## 1 rs2476601 0.8640309
## 2 rs6679677 0.1359691
## 
## $Claimed
## [1] 1
## 
## $Corrected
## [1] 0.8035921
## 
## $nSNPs_in_region
## [1] 639

2. Build 36 method

  • Get genotype data from corresponding region from share/Projects/abc/input_files/refdatasets/datafilesD/ –> chr1:113619999-114460000 and convert the SNP names to rs IDs. Remove variants whose names cannot be matched (583 SNPs left).
  • Get SNPs in this region from ic_t1d_onengut_allresults.csv file (640 SNPs).
  • Remove the columns from \(X\) representing SNPs which do not appear in both of the above (520 SNPs left). The dimensions of X is 9670*520.
  • Get LD matrix. LD <- cor2(x)
  • Proceed as normal.
## $Credset
##     name.rs        pp
## 1 rs2476601 0.8640309
## 2 rs6679677 0.1359691
## 
## $Claimed
## [1] 1
## 
## $Corrected
## [1] 0.889136
## 
## $nSNPs_in_region
## [1] 520

Comparing Build Methods

  • Find the SNPs which are used in both the above methods (495 the same).

  • Proceed as normal for both methods.

  • Do they both get stuck in a loop? Yes.

Build 37 Method:

Build 36 Method:

## $build37_claim
## [1] 1
## 
## $build37_corr
## [1] 0.882864
## 
## $build37_reqthr
## [1] 1
## 
## $build37_newCS_claim
## [1] 1
## 
## $build37_newCS_corr
## [1] 0.989864
## 
## $build36_claim
## [1] 1
## 
## $build36_corr
## [1] 0.8977281
## 
## $build36_reqthr
## [1] 1
## 
## $build36_newCS_claim
## [1] 1
## 
## $build36_newCS_corr
## [1] 0.989864